Development of generalized two-phase (oil/gas) and three-phase relative permeability predictors using artificial neural networks.

摘要

Due to the highly non-linear nature of multi-phase flow dynamics in porous media, relative permeability is one of the foremost vaguely understood phenomena in fluid flow transport. At the same time, in no uncertain terms, relative permeability is one of the most important rock-fluid properties required almost in all calculations of multi-phase flow dynamics in porous media.; Laboratory determination of relative permeability characteristics is labor intensive and can be complicated. Empirical models to predict relative permeabilities based on rock and fluid properties have experienced relatively mediocre success. The difficulties of experimental measurement of relative permeabilities and the limited success of empirical models justify the need for an alternative tool to estimate relative permeability characteristics.; Artificial Neural Network (ANN) technology has been utilized in a variety of applications ranging from pattern recognition to optimization protocols. ANNs perform non-linear, multi-dimensional interpolations making it possible to capture the non-linear relationships between the input and output parameters. In this way, it has a potential to identify some of the vague non-linear relationships that control the relative permeability characteristics.; In this study, two-phase (oil/gas) and three-phase relative permeability predictors are developed using backpropagation networks. In this category of networks, information is passed from input layer to output layer, and calculated errors are propagated back to adjust the connection weights in a sequential manner to improve the predictive capabilities of the models. In the development of the models, some of the experimental relative permeability data sets along with some commonly reported rock and fluid properties obtained from the literature are used during the training stage, while other sets are preserved to test the prediction ability of the models. The two-phase (oil/gas) relative permeability models are found to perform in a satisfactory manner within a wide spectrum of basic rock and fluid properties. Similarly, three-phase relative permeability models are found to have good predictive capabilities in accurately producing the missing or additional three-phase relative permeability values for the training data sets. Furthermore, they are found to be capable of effectively predicting the three-phase relative permeability values at various saturation combinations for unknown systems with different rock and fluid properties.